B cells are the immune cells responsible for creating antibodies, and most B cells, known as B2 cells, produce antibodies in response to a pathogen or a vaccine, providing defense and immunity against infections. But a small subset of long-lived B cells, known as B1 cells, are quite different from their short-lived cousins, the B2 cells.
Instead of producing antibodies in response to invaders, they spontaneously make antibodies that perform vital housekeeping functions, such as removing waste like oxidized LDL cholesterol from the blood.
Like all the cells in the body, B1 and B2 cells have the same DNA, and therefore the same starting set of instructions. It is through epigenetic modifications, which open and close different areas of the genome to the machinery that reads the genetic instructions, that the same genome can be used to create unique instructions for each cell type.
Understanding how the different epigenetic landscapes – the changes in instructions – allows for these differences in such similar cells is both an important fundamental question in immunology and can help scientists better understand diseases linked to B cells’ dysregulation.
Shiv Pillai, MD, Ph.D., a core member of the Ragon Institute of MGH, MIT and Harvard, studied the DNA modifications present in both cell types during different stages of development to identify an epigenetic signature that may determine whether a cell becomes a B1 or a B2 cell. This work was published recently in the journal Nature Communications.
“Through our analysis, we found the fate of a B cell is determined by epigenetic modifications driven by a protein called DNMT3A,” says Vinay Mahajan, MD, Ph.D., an instructor in Pathology at the Ragon Institute and the paper’s first author. “Genetic studies in humans link the genomic regions with these markers to a variety of immune-mediated disorders.”
The team studied CpG methylation, a type of epigenetic modification that opens up specific areas of DNA and marks regulatory elements that can turn genes on or off. They discovered a set of regulatory elements with unique features in these B1 and B2 cells.
In most cases, CpG methylation is permanent and, once added, is even passed on when the cell replicates. But in B cells, the protein DNMT3A had to continually work to maintain these epigenetic modifications. If DNMT3A was removed from B1 cells, the epigenetic modifications were lost, and chronic lymphomic leukemia (CLL), a cancer caused by the uncontrolled replication of B1 cells, would arise.
“The antibodies they create help prevent clots and heart attacks. At the same time, understanding what genetic factors regulate them can help us better understand what happens when their regulation goes awry and leads to CLL and other diseases.”
DNA methylation is one of the major epigenetic mechanisms that critically influence gene expression, genomic stability, and cell differentiation1,2,3. In mammals, DNA methylation predominantly occurs at the C-5 position of cytosine within the symmetric CpG dinucleotide, affecting ~70–80% of the CpG sites throughout the genome4.
Mammalian DNA methylation patterns are mainly generated by two de novo DNA methyltransferases, DNMT3A and DNMT3B5. The catalytically inactive DNMT3-like protein (DNMT3L) has an important regulatory role in this process by acting as cofactor of DNMT3A or DNMT3B6,7,8.
In addition to CpG methylation, DNMT3A and DNMT3B introduce non-CpG methylation (mainly CpA) in oocytes, embryonic stem (ES) cells, and neural cells4,9,10,11. The presence of non-CpG methylation was reported to correlate with transcriptional repression12,13 and the pluripotency-associated epigenetic state4,14, lending support for non-CpG methylation as an emerging epigenetic mark in defining tissue-specific patterns of gene expression, particularly in the brain.
DNMT3A and DNMT3B are closely related in amino acid sequence5, with a C-terminal methyltransferase (MTase) domain preceded by regulatory regions including a proline–tryptophan–tryptophan–proline (PWWP) domain and an ATRX–DNMT3–DNMT3L-type (ADD) zinc finger domain15,16.
Previous studies have indicated a partial redundancy between the two enzymes in the establishment of methylation patterns across the genome5,17; however, a single knockout (KO) of either DNMT3A or DNMT3B resulted in embryonic or postnatal lethality, indicating their functional distinctions5,17,18,19.
Indeed, it was shown that DNMT3A is critical for establishing methylation at major satellite repeats and allele-specific imprinting during gametogenesis8,17, whereas DNMT3B plays a dominant role in early embryonic development and in minor satellite repeat methylation5,17. Mutations of DNMT3A are prevalent in hematological cancers such as acute myeloid leukemia (AML)20 and occur in a developmental overgrowth syndrome21; in contrast, mutations of DNMT3B lead to the
Immunodeficiency, centromeric instability, facial anomalies (ICF) syndrome5,22,23,24. Previous studies have indicated subtle mechanistic differences between DNMT3A and DNMT3B9,25,26,27,28, including their differential preference toward the flanking sequence of CpG target sites29,30,31,32. However, due to the limited number of different substrates investigated in these studies, global differences in substrate recognition of DNMT3A and DNMT3B remain elusive.
Our recently reported crystal structure of the DNMT3A–DNMT3L heterotetramer in complex with CpG DNA33 revealed that the two central DNMT3A subunits bind to the same DNA duplex through a set of interactions mediated by protein motifs from the target recognition domain (TRD), the catalytic core and DNMT3A–DNMT3A homodimeric interface (also called RD interface below). However, the structural basis of DNMT3B-mediated methylation remains unclear.
To gain mechanistic understanding of de novo DNA methylation, we here report comprehensive enzymology, structural and cellular characterizations of DNMT3A- and DNMT3B-DNA complexes. Our results uncover their distinct substrate and flanking sequence preferences, implicating epigenomic alterations caused by DNMT3 mutations in diseases. Notably, we show that the catalytic core, TRD domain and RD interface cooperate in orchestrating a distinct, multi-layered substrate-readout mechanism between DNMT3A and DNMT3B, which impacts the establishment of CpG and non-CpG methylation patterns in cells.
reference link: https://www.nature.com/articles/s41467-020-17109-4
More information: Vinay S. Mahajan et al, B1a and B2 cells are characterized by distinct CpG modification states at DNMT3A-maintained enhancers, Nature Communications (2021). DOI: 10.1038/s41467-021-22458-9